Examining Early Diagnostic and Lifestyle Intervention Strategies for Alzheimer‘s disease

Kathryn Ellis, PhDNational AIBL study coordinator

Research Fellow, University of Melbourne

www.aibl.csiro.au

What is the AIBL study?

• Large scale cohort study launched in September 2006

• 3 year prospective longitudinal study of Alzheimer’s disease

• 1000+ participants (minimum age 60 years)

• Patients with AD, MCI and healthy volunteers

• Multi-disciplinary approach, 4 research streams – cognitive, imaging, biomarkers and lifestyle

• Focus on early detection, towards lifestyle interventions

AIBL is a collaborative venture

• CSIRO P-Health initiative, bringing together a “cluster” of Australian researchers and research organisations

• CSIRO P-Health*• University of Melbourne*• Neurosciences Australia Ltd (NSA)*• Edith Cowan University (ECU)*• Mental Health Research Institute (MHRI)*• National Ageing Research Institute (NARI)• Alzheimer’s Australia• Austin Health• University of WA (UWA)• CogState Ltd.• Charles Gairdner Hospital Radiology and Nuclear Medicine

*denotes signatories to the AIBL study contract

• Multidisciplinary, collaborative, brings together researchers in Perth, Adelaide, Melbourne, Canberra and Brisbane (40% subjects from WA, 60% from Vic)

Basis of AIBL

• Treatment strategies are likely to be most effective if administered early

• Development of lifestyle interventions to prevent or delay AD onset would have significant effects at the population level

• Currently no certain method of detecting whether a person is likely to develop AD.– Mild cognitive impairment (MCI): 30 – 60% develop AD

• For treatment to be most effective we must be able to identify who is at risk of developing AD years before symptoms develop – this is what AIBL seeks to do

Study aims

1. To improve the understanding of the pathogenesis and diagnosis of Alzheimer’s disease using neuropsychological, neuroimaging and biomarker techniques, with a focus on early diagnosis of AD

2. To examine lifestyle and diet factors that may be involved in the pathogenesis of AD, towards future lifestyle intervention

Volunteers

• We sought to recruit 1000+ individuals – At least 200 individuals with AD (McKhann et al., 1984)

– At least 100 individuals with Mild Cognitive Impairment (MCI) (Petersen et al., 1999; Winblad et al., 2004).

– At least 700 healthy individuals without cognitive impairment. This group included:

• Volunteers with at least one copy of the ApoE ε4 allele, • Volunteers without a copy of the ApoE ε4 allele, • Volunteers who expressed subjective concern about their

memory function

Blood taking procedure

• Fasting samples are taken and participants receive breakfast following collection

• Approx 80 ml blood is collected in 12 tubes

Clinical Pathology

80ml blood

ESR and FBE HomocysteineSerum and red cell folateB12GlucoseCeruloplasminFerritin/Transferrin/IronInsulinOestradiolLHThyroid (TSH,FT4,FT3)PSA (MALES ONLY)

Neuropsychology• Memory/learning: California Verbal Learning Task (CVLT-II)

Logical memory II (WAIS)

• Language: Controlled Oral Word Association (COWAT)Boston Naming Test (BNT)

• Working memory: Digit span task

• Attention: Digit Symbol (WAIS III)

• Executive function: Stroop

• Spatial: Complex figure of Rey

• CogState battery: www.cogstate.com

Other clinical measures

• Hospital Anxiety and Depression Scale (HADS)• Geriatric depression scale (GDS), 15 item version • Clinical dementia rating (CDR) • Wechsler Test of Adult Reading (WTAR)• Mini-Mental State Examination (MMSE)• Informant Questionnaire of Cognitive Decline in the Elderly (IQCode)

SUMMARY• Comprehensive assessment, takes 2 hours• Neuropsychologists suspend cognitive testing is participant/patient

is overwhelmed (although this is rare)• Clinical review panel meets monthly to review diagnostic

classification of all AD and MCI cases and all healthy controls with any results of interest or concern

Aβ Neuroimaging

• The structure of this protein was first described by a team co-led by Professor Colin Masters (1985)

• Aβ can now be imaged in the living brain by PiB PET

• 25% of volunteers to receive PET Pitsburgh Compound B (PiB) scans

Lifestyle assessment

Full cohort– Food Frequency Questionnaire (FFQ)– Physical Activity Questionnaire (IPAQ)

Sub-group– Actigraph Uni-axial Accelerometer

• Small devices (size of matchbox), 7 days per volunteer

– Dual Energy X-Ray Absorptiometry (DEXA)• “Gold standard” body composition assessment, low dose

radioactivity scan

Current status of study

BASELINE FIRST FOLLOW UP (18mth)

Clinical/cognitive data

80ml blood

Lifestyle information

PET & MRI scans

Clinical/cognitive data

80ml blood

Lifestyle information

PET & MRI scans

TIMELINE

Current status

Cohort at baseline

337920

211AD

133MCI

At Presentation/ Pre-assessment

BASELINE COHORT

(1112 participants)

Cohort sub-groups

180 NINCDS-ADRDAProbable AD

31 NINCDS-ADRDAPossible AD

395 Memory Complainers

372 Non Memory Complainers

3746745

768Healthy controls

54 Excluded/withdrawn

823Healthy controls

150MCI

193AD

49 Amnestic-single domain

77 Amnestic-multi domain

1 Non-amnestic-single domain

6 Non-amnestic-multi domain

1718

39 / 6144 / 5643 / 57Sex (%m/ f)

64.447.526.8ApoEε4 carriers (%)

0.90.50.0CDR

19.126.228.9MMSE

78.5 (8.6)76.2 (7.6)70.5 (7.0)Age

211133768N

ADMCI HC

Cohort characteristics

0

10

20

30

40

50

60

70

80

CVLT learning

CVLT-II learning T-scores

Fig 2: CVLT II recall

-3

-1

1

3

1 2

CVLT-II recall following short (left) and long (right) delay, displayed as Z scores

Fig 1: CVLT II learning

HC MCI AD

SHORT LONG

Memory measures

56789

1011

1213

Dig it Sp an Dig it Symb o l Cod ing

WAIS-III Digit Span and Digit Span Coding Age Scaled Scores

0

5

10

15

FAS Animals + BoysNames

Fruit/FurnitureNaming

Fruit/FurnitureSw itching

Letter and Category Fluency Age Scaled Scores

Verbal FluencyAttention & Processing Speed

Other cognitive tasks

• Participants (n=287)• 177 Healthy elderly controls (HC)

81 Non-memory complainers (49% ApoE ε4+ve)96 Memory complainer (40% ApoE ε4+ve)

• 57 MCI participants• 53 AD participants

Neuroimaging sub-group

74%52%43%% ApoE ε4

24/3029/2987/90Sex (M/F)

20.5 ± 4.927.1 ± 2.328.8 ± 1.2MMSE

74.0 ± 8.777.4 ± 7.573.6 ± 7.6Age

ADMCI HC

Representative images

SUVR3.0

1.5

0.0

AD

HC

29%64%

98%

HC MCI AD

PiB +ve or “AD-like”

PiB - ve or “HC-like”

29%64%

98%

HC MCI AD

PiB +ve or “AD-like”

PiB - ve or “HC-like”

Significant differences between the three groups (p<0.001)

Percentage of PiB+ volunteers

HC MCI AD

PiB +ve or “AD-like”

PiB -ve or “HC-like”

In the HC subgroup, 49% of HC ApoEε4+ were PIB+ve, compared to 21% of the HC ApoEε4-ve group.

PiB +ve or “AD-like”

PiB - ve or “HC-like”

ε4- ε4+

51%79%

21%49%

Healthycontrols: ApoE ε4

0

5

10

15

20

25

PiB -ve PiB +ve

Healthy volunteers

23 PiB -ve subjects scanned,

only 1 converted to MCI (4%)

10 PiB +ve subjects scanned,

3 converted to MCI/AD (30%)

15 PiB+ve subjects scanned,

12 converted to AD (80%)

14 PiB-ve subjects scanned,

5 converted to dementia (36%)

MCI patients

0

2

4

6

8

10

12

14

PiB -ve PiB +ve

Predicting conversion (pre-AIBL data)

PiB -ve PiB +ve PiB +vePiB -ve

Rowe & Villemagne 2009

• 553 Healthy Volunteers (57% female, mean age 70 + 8.5) completed the IPAQ (International Physical Activity Scale)

• IPAQ scores were correlated with plasma Aβ levels (ratio Aβ42/ Aβ40)

Exercise and Blood plasma Aβ

0.19

0.195

0.2

0.205

0.21

0.215

0.22

0.225

0.23

1 2 3 4

Aβ

42

/Aβ

40

Ra

tio

Quartiles1 = lowest activity levels; 4 = highest activity levels

Brown et al. 2009

MEASURE OF FOOD INTAKE• Cancer Council of Victoria Food frequency questionnaire (CCVFFQ)

- Total consumption of foods and nutrients- Consumption of fish (Omega 3, DHA)

CLINICAL/NEUROPSYCHOLOGICAL MEASURES• Mini-mental state exam (MMSE)

- The most widely used brief measure of cognitive function.

• CVLT II- Primary memory measure- Consists 16 word list (read 5 times)- Measures learning, short recall and long recall

ANALYSIS• Healthy controls (639 volunteers)• Partial correlations were performed between nutritional & cognitive variables

Diet and cognitive function

Greater fish consumption was associated with higher cognition in healthy controls

Nutrient intake (diet only)

NEUROPSYCHOLOGICAL MEASURE (RAW)

Total Fish

Total Omega 3

Omega 3/6 ratio

DHA

MMSE P=0.012*

0.013* 0.047* 0.013*

Learning (T score) 0.001** 0.000** 0.046* 0.000**

Short term recall/ Retention 0.031* 0.035* 0.070 0.033*

Delayed recall 0.009** 0.010** 0.076 0.010**

Recognition 0.030* 0.046* 0.016* 0.050*

*Correlation is significant at the 0.05 level (2-tailed); **Correlation is significant at the 0.01 level (2-tailed). n=639 Healthy controls (memory complainers + non-memory complainers). Omega 3 = ALA + DHA +EPA. Controlled for sex, years of education, genotype, BMI and age

Martins, Ward et al. 2009

• 50% of follow-ups completed. On track for full completion by December 2009

• Follow-up classifications for 432 people have been confirmed

N/A001717AD

4N/A22228MCI

28N/A377387HC

Reclassified AD

Reclassified MCI

Reclassified HC

Unchangedfrom baseline

Number retested

Baseline classification

Follow-up (18mth): Progress to date

• The baseline characteristics demonstrate a highly motivated, well balanced cohort.

• Neuropsychological findings reveal HC>MCI>AD on performance, as expected

• Brain imaging: – AD showed marked cortical and striatal PiB retention with

relative sparing of occipital and sensorimotor cortex.– 67% of MCI participants showed significant cortical PIB

retention, similar to AD.– 36% of HC volunteers showed significant cortical PIB retention,

though lower than in AD– Greater PiB burden in ApoE ε4+ volunteers than ApoE ε4-

volunteers

Summary

• Levels of exercise correlate with blood amyloid levels

• Higher intake of fish is correlated with better cognition

• First follow-up is 50% complete

• These data will assist development of techniques for early detection of AD and provide a cohort suitable for targeted early intervention studies

Summary - 2

Austin Hospital, Melbourne WA PET Centre, PerthSir Charles Gairdner Hospital, PerthEdith Cowan University CSIRO ICT Centre, Brisbane University of MelbourneMental Health Research Institute Neurosciences Australia Ltd National Ageing Research InstituteUniversity of WA (UWA)CogState Ltd.

Acknowledgements and thanks

• The AIBL volunteers and their families

• The staff of our collaborating organisations

Prof. David Ames (NARI)Mr. Lindsay Bevege (NSV)Dr. Kathryn Ellis (Uni of Melb.)Prof. Peter Hudson (NSV) Prof. Ralph Martins (ECU)Prof. Colin Masters (MHRI)Dr. Andrew Milner (NSV)Prof. Chris Rowe (ARMC) Dr. Cassandra Szoeke (CSIRO)

Management committee

Osca AcostaDavid AmesJennifer AmesManoj AgarwalDavid BaxendaleKiara Bechta-MettiCarlita BevageLindsay BevegePierrick BourgeatBelinda BrownAshley BushTiffany CowieKathleen CrowleyAndrew CurrieDavid DarbyDaniela De FazioDenise El- SheikhKathryn Ellis Kerryn DickinsonNoel FauxJonathan FosterJurgen FrippChristopher FowlerVeer Gupta Gareth Jones

Jane Khoo Asawari Killedar Neil KilleenTae Wan Kim Eleftheria KotsopoulosGobhathai KunarakRebecca LachovitskiNat LenzoQiao-Xin Li Xiao Liang Kathleen LucasJames LuiGeorgia MartinsRalph Martins Paul MaruffColin MastersAndrew MilnerClaire MontagueLynette MooreAudrey MuirChristopher O’HalloranGraeme O'KeefeAnita PanayiotouAthena PatonJacqui PatonJeremiah Peiffer

Svetlana PejoskaKelly PertileKerryn Pike Lorien PorterRoger PriceParnesh RanigaAlan Rembach Miroslava RimajovaElizabeth RonsisvalleRebecca Rumble Mark RodriguesChristopher RoweOlivier SalvadoJack SachGreg SavageCassandra SzoekeKevin TaddeiTania TaddeiBrett TrounsonMarinos Tsikkos Victor VillemagneStacey Walker Vanessa WardMichael WoodwardOlga Yastrubetskaya

AIBL study team